Project Summary/Abstract Tuberculosis (TB) is a widespread bacterial infectious disease that kills nearly 1.5 million people annually. While effective drug therapy for TB has been available for more than 50 years, there is a substantial number of drug resistant clinical cases that are signi?cantly impacting public health. Drug regimens for TB are designed to limit the emergence of resistance by using multiple drugs concurrently (combination therapy) which greatly increases the time and cost of their development. While the U.S. Food and Drug Administration (FDA), in partnership with the recently formed Critical Path to New TB Drug Regimens (CPTR) initiative, now provides regulatory guidance for developing new drug combinations as a single unit, and while several new anti-TB regimens are in clinical testing under this FDA guidance, there are critical questions about how to establish the optimal dose of each individual drug within these new combination regimens. Dosage regimens for new anti-TB drug combinations are generally based on ?nding an optimal dose for every single drug in the preclinical stage, and through Phase II dose-ranging clinical trials. While tailoring the doses of each individual drug within a drug combination could potentially yield a more effective and better tolerated treatment regimen, the exponential increase in the in vitro methodologies, animal ef?cacy studies, and clinical testing required to identify such doses for combinations of three or more drugs needed for TB would be prohibitively expensive. To address this gap in TB drug development we propose a new approach to dosage regimen design of combination drug therapies that consists of (1) the use of conventional preclinical and clinical measurements to inform a mathematical dose-response model for a speci?ed drug combination in TB patients, (2) the integration of this mathematical model with a biologically inspired genetic algorithm to design dosage regimens in a manner analogous to natural selection, and (3) the empirical evaluation of these optimized regimens in experimental TB-infection models. To establish our approach with a clinically relevant example, we will design optimized dosage regimens for the new anti-TB combination pretomanid + moxi?oxacin + pyrazinamide (PaMZ); a promising and urgently needed treatment option for patients with multidrug resistant (MDR) TB, currently assessed in a Phase II clinical trial. There is a large amount of high quality preclinical and clinical data for this TB drug combination that will provide a sound evidence base to develop our computational framework and to test our conclusions. Successful completion of the proposed aims will establish new methods and tools to better translate preclinical studies to clinical dosage regimen design for future anti-TB combinations. While motivated by the needs of TB drug development, this project includes innovations that apply to the treatment of other diseases such as cancer, human immunode?ciency virus (HIV) infection, and malaria.